A blueprint for local and distal glioblastoma invasion programs

Glioblastoma (GBM) invasion into brain parenchyma presents significant challenges for treatment but remains poorly understood. In this study, we combined single-cell RNA sequencing (scRNA-seq), spatial transcriptomics, and multiplexed imaging of orthotopic xenograft models to investigate GBM invasion. We first screened 20 patient-derived gliomasphere models for their distal (i.e., extending to the contralateral hemisphere) and local invasive potential in mice. We found that models with distal invasion potential are enriched with oligodendrocyte progenitor-like (OPC-like) cells, while models with only local invasion potential are enriched with mesenchymal-like (MES-like) cells. These patterns reflect predominantly peri-axonal vs peri-vascular invasion routes, respectively. Next, we analyzed the transcriptomes of invading cells within models (compared to tumor core) and identified novel programs associated with distal and local invasion. Thus, we decouple transcriptional features associated with invasion potential from those associated with the process of invasion. We validated our findings by spatial transcriptomics and multiplexed imaging, further describing the spatial niche of invasive cells. Taken together, our results provide a blueprint for the invasive potential of GBM cell states and of the programs associated with invasion across different scales. Overall design: We developed a pooled in vivo system to simultaneously analyze the contralateral invasion capacity of multiple patient-derived GBM models. In this assay, we grew 20 GBM patient-derived gliomasphere models separately and pooled them for stereotaxic injection into the mouse brains (10 models per pool – [Pool #1, Pool #2] with 2 mice replicates). At symptom onset, the mice were euthanized, and brains were dissected under a stereomicroscope and separated into the primary injection site and the contralateral hemispheric invasion fractions. Malignant cells from each site were isolated by flow cytometry based on GFP expression and profiled by scRNA-seq using the 10X Genomics 3' gene expression platform.To characterize the program expressed by locally-invading glioma, we employed fluorescent-guided microdissection and isolated around 200 malignant cells associated with vasculature within the ipsilateral hemisphere of two non-distally invasive models (MGG65 and MGG123). These cells were profiled by scRNA-seq using SS2, along with around 200 cells of the tumor core of each model.